01 5.3/Tahoe..

[Chuck L.]

I'm chasing a loss of fuel psi, when key is turned off. [130,000 mi vehicle.]
So far, I've found a couple injs leaky, but not serious enuf to cause an immediate drop to near zero...[<2 mins and it's over!].
Thus far:
1. I've cleaned the injectors. Were NASTY!
2. Replaced the fp regulator.
3. Replaced the filter.
I'm thinking it's down to the check valve in the pump, has taken a dump.
The psi at/near 0, key on, engine off, it takes 2-3 cycles to get psi up to op pressure. If I go directly to start, it takes considerable time to raise the psi far enuf to start.

I've scanned it w/ Flashscan.. no codes, and all parameters look good.
I've seen some postings on other boards, about GM truck pumps being a problem....electrical or??
Can someone confirm the exsistence of the check valve, B4 I drop the tank?

[joecar]

Chuck,

If you replaced the FPR, then the problem probably is the check valve... the check valve would be integral with the pump itself...

I don't know if all models have a check valve... but the pump itself will not hold pressure when not running.

but with that many miles the pump/motor will be worn anyway (on their way out) and would need replacement soon.

If you have access to an oscilloscope you can plot the fuel pump current using a current clamp... you will see a repeating series of humps... if any humps look irregular/flat then the pump motor is dying (problem with that winding or commutator segment)...

this waveform also lets you determine the rpm of the pump... if the rpm is low, then the motor is laboring (bad motor, bad pump, restricted flow/filter...

If you can justify a scope, then it will reveal many things... unfortunately all scopes are quite expensive.

Regards
Joe

[ChipsByAl]

There is a check valve in the pump, however it is not user replaceable. Get yourself a new GM fuel pump. Depending on model it may come with a new design connector as well. 130k is a long life for a pump, its due. Get a new fuel filter while you are at it.
Al

[Chuck L.]

Thanks, Joe and Al.. DFE1 and I have been looking at a small o'scope, for general use, and with the injection test bench..
A new pump it is!

[Chuck L.]

The 01 has a replacement that is the entire setup..hanger, sender, pump.. All for the "paltry sum" of near $600!!
I think the customer has gone into "coronary arrest"...I haven't heard back from him, since I gave him the "news".
Joe, here's a scope that DFE and I are looking at.. NE comment[s] are appreciated.

http://www.apogeekits.com/pc_oscilloscope_pcsu1000.htm

TIA,

[joecar]

Does your customer realize this: how much repair money has he spent on his car in the last 130,000 miles...? Probably very little.

[Chuck L.]

Does your customer realize this: how much repair money has he spent on his car in the last 130,000 miles...? Probably very little.

His car, truck, and the farm equipment he owns, are all on a "if it's broke fix it" basis. he's now at the "fix it" point w/ his car!

what do you think of the scope I posted about? Would it be a reasonably priced tool??

[dfe1]

His car, truck, and the farm equipment he owns, are all on a "if it's broke fix it" basis. he's now at the "fix it" point w/ his car!

what do you think of the scope I posted about? Would it be a reasonably priced tool??

Get one that can also be used for rectal exams. That way, you can help some of your "customers" locate their brains.

[joecar]

Chuck,

That looks like a well priced easy to use scope...

But here are my concerns (see the cross-reference comparison pdf right-hand column):

1. Injector turn-off inductive-kick voltage can be reach 100-200V...
the PCSU1000 is rated for 30V (with the x10 probe)...
you will need a x50 or x100 probe (if there is one)...
or a x5 or x10 attenuator plus a x10 probe.

2. the PCSU1000 has a per channel memory length 4096 samples (at resolution 8 bits/byte and 1 byte/sample)...
at 600 rpm (10 rev/s) there are 5 complete cycles...
an injector fires once every cycle, so the cycle period is 200 ms/cycle...
if you display two cycles (400ms) across the screen, the best resolution will be 4096 samples for 400 ms,
or each 1 ms has 10.24 samples (round this to 10)...
10 samples in 1 ms is marginal to accurately see the injector turn on/off slopes...
it might be ok... it would be ok if you view less than a complete cycle (1/2 cycle would contain 40 samples);
I would have liked more samples... (i.e. a longer memory buffer).


Ask the scope vendor/maker about those 2 points...
explain your intended application (measuring injector waveform slopes) and see what he says...

see if he can provide a bigger memory buffer...

see if he will let you trial the scope before buying...

and make sure it comes already assembled, ready to go.

Cheers
Joe

[joecar]

Here's a waveform from my Pico scope taken from injector on #1.


For ignition coil secondary events (using a kV probe) you need a much deeper memory...
you want to see 1 microsecond ripples in a 2+ millisecond event.